addressing the ''research gap' in special education...

Addressing the ''Research Gap'in Special Education ThroughMixed Methods

Learning Disabilit/ Quarterly34(3)208-218© Council for Learning Disabilities 2011Reprints and permission: http://www.sagepub,com/journalsPermissions,navDOI: 10,1177/07319487114I7S59http://tdq,sagepub,com

tSAGE

Janette K. Klingner' and Alison G. Boardnnan'

Abstract

At least some of the challenges faced in special education, such as the disproportionate representation of culturally andlinguistically diverse students, the gap between research and practice, and inequitable educational opportunities, can beexplained in part by a research gap, or, in other words, a failure to conduct the different types of research best suitedfor addressing the complicated issues faced in schools. In this article we discuss the benefits of being more open to andwelcoming of mixed methods when conducting special education research. We provide an overview of mixed-methodsresearch and explain different philosophical concepts associated with mixed methods. We emphasize why it is importantto foreground culture when conducting educational research. We also compare educational research with research inthe medical field and challenge the notion of randomized controlled trials as the "gold standard." We finish by sharing anexample of our own mixed-methods research.

Keywords

mixed methods research, research to practice, qualitative research

At least some of the challenges faced in education, such asthe achievement gap between White students and studentsof color (Lee, 2002), the disproportionate representation ofculturally and linguistically diverse students in special edu-cation (Donovan & Cross, 2002; Harry & Klingner, 2006),the gap between research and practice (Elmore, 1996;Schneider & McDonald, 2007; Vaughn, Klingner, &Hughes, 2000), and inequities in educational opportunities(Ball & Forzani, 2007; da Silva, Huguley, Kakli, & Rao,2007), can be explained in part by a research gap, or, inother words, a failure to conduct the different types ofresearch best suited for addressing the complicated issuesfaced in schools. A broader view of research that encom-passes mixed methods would enable special educationscholars to collect empirical data relevant to issues involvingculture, language, social interaction, and cognition (Gee,2001), thereby expanding the kinds of research problemsthat can be addressed and the applicability of findings. Ourposition is that the U.S. Department of Education overrelieson randomized control trials for determining what works(Chatterji, 2005; Eisenhart, 2006; Raudenbush, 2005). Atthe very least, the practicability of randomized control trialsis limited when the data collected are insufficient for contex-tualizing the results or for supporting real world applications(Spillane et al., 2010). Researchers sometimes gather andanalyze a wealth of qualitative data that are never reported orused to understand quantitative findings because of the

premium placed on significance values and effect sizes. Weclaim that mixed methods that combine quantitative andqualitative research tools can support stronger scientificinferences than when either is employed in isolation (Feuer,Towne, & Shavelson, 2002) and that mixed methods are bet-ter positioned to determine what works (Creswell & PianoClark, 2011). On this point, Raudenbush (2005), a highlyrespected quantitative research methodologist, asserts.

The question before us now is not whether to employ mixedmethods in education research generally; rather, the questionis how to employ them in the service of a newly dominantresearch agenda that seeks to evaluate claims about thecausal effects of interventions aimed to improve teachingand leaming in the nation's classrooms. . . . Well-designedrandomized experiments are, I believe, necessary but notsufficient for determining what works, (p. 25)

Mixed-methods designs are better suited to unravelingeducational phenomena "of enormous complexity" (Berliner,2002, p. 20). "Because the U.S. education system is so

'University of Colorado at Boulder, Boulder, CO, USA

Corresponding Author:Janette Klingner, University of Colorado at Boulder, 50 Buckthorn Drive,Littleton, CO 80127Email: [email protected]

Klingner and Boardman 209

heterogeneous and the nature of teaching and leaming socomplex, attention to context is especially critical for under-standing the extent to which theories and findings may gen-eralize to other times, places, and populations" (Shavelson &Towne, 2002, p. 5). Educational programs are implementedin real world contexts, in authentic schools, with characteris-tics that are both unique to a particular context and sharedacross contexts. Mixed-methods research can help to estab-lish cross-context pattems of regularity and determine uniquewithin-site variables (Greene, 2005). Whereas quasi-experi-mental and experimental approaches provide informationabout which instmctional methods are most effective in ageneral sense, qualitative research helps establish whichcontextual factors may influence the success of an approach.Mixed methods can lead to insights about possible chal-lenges to implementation as well as the circumstances underwhich a practice is most likely to be successful, adding depthand breadth not available through quantitative designs alone.We propose that it is not enough to ask, "What works?"Researchers must also ask, "What works with whom, bywhom, in what contexts, under what circumstances, and forwhat purposes?" (Klingner & Edwards, 2006). Mixed-methods research can answer questions about why a particu-lar approach works for some children and not others(Raudenbush, 2005). For example, by observing teachers'implementation of a practice and interviewing teachers andstudents, researchers can begin to produce plausible expla-nations for variations in quantitative student outcome data,new hypotheses, and ideas for fine-tuning the practice, aswell as for refining data collection and analysis methods.Such an approach can lead to more nuanced understandingsof how school and neighborhood contexts, as well as lin-guistic and cultural variables, influence teacher implementa-tion and student outcomes.

In this article we discuss the benefits of being more opento and welcoming of mixed methods when conducting spe-cial education research. We provide an overview of mixed-methods research and explain different philosophicalconcepts associated with mixed methods. We emphasizewhy it is important to foreground culture when conductingeducational research and compare educational research withresearch in the medical field. We also share an example of ourown mixed-methods research.

An Overview, ofMixed-Methods ResearchTo generate a basic definition of mixed-methods research,Johnson, Onwuegbuzie, and Tumer (2007) asked numerousleaders in mixed-methods research for their definitions ofthe approach and compiled their responses. They concludedthat mixed-methods research is a combination of "elementsof qualitative and quantitative research approaches (e.g., useof qualitative and quantitative view points, data collection.

analysis, inference techniques) for the purposes of breadthand depth of understanding and corroboration" (p. 123).Mixed-methods research legitimizes the use of multipleapproaches in answering research questions. It is inclusive,pluralistic, complementary, and eclectic.

Purposes of Mixed Methods

Mixed-methods research has several purposes and can addressmany types of research questions. The emphasis shouldalways be on figuring out the most appropriate methods toaddress a specific purpose and answer particular questions.Mixed methods can illusfrate and explain quantitative find-ings, describe both process and product, check reasons forunexpected effects, develop the basis for instmments, showthe extent of generality, validate and friangulate other data,and fulfill social or political purposes (Krathwohl, 2009).There are several reasons for adding qualitative data before,during, and after intervention studies (adapted from Creswell& Piano Clark, 2011, p. 192):

Reasons for adding qualitative data before beginningthe intervention;

• To develop an instrument for use in the study• To understand the participants, context, and environ-

ment• To document a need for the intervention• To compile comprehensive baseline information

Reasons for adding qualitative data during theintervention:

• To validate quantitative outcomes with qualitative datarepresenting participants' perspectives

• To understand the impact of the intervention on partici-pants

• To understand different characteristics of the partici-pants that might affect outcomes

• To understand barriers and facilitators to the implemen-tation of the intervention

• To understand participants' experiences during the inter-vention

• To understand factors in the environment that mightpotentially affect the outcome of the intervention

• To check the fidelity of implementation procedures (andto understand why fidelity might be low)

• To identify potential mediating and moderating factors

Reasons for adding qualitative data after concludingthe intervention:

• To understand participants' perspectives on the results of thestudy

210 Learning Disability Quarterly 34(3)

• To elicit participant feedback that can be used to revisethe intervention

• To help explain quantitative outcomes• To determine the long-term effects of the intervention• To determine the extent to which teachers continue

using an intervention• To understand in more depth how the components of the

intervention worked in a theoretical model• To assess the context when comparing outcomes

Types of Mixed-Method Designs

Mixed-methods research can be designed in different con-figurations that vary depending on the order in which dataare collected and which form of data collection is fore-grounded. Creswell and Piano Clark (2011) described sixpossible designs. We list these designs here, along withsample research questions.

Convergent Design: Quantitative and qualitative dataare collected concurrently, analyzed separately,and then merged.

To what extent do quantitative and qualitative find-ings converge?

Example: To what extent do self-esteem survey ratingsagree with the views of middle sehool boys withleaming disabilities (LD) about their self-esteem?

Explanatory Design: Quantitative data are collectedand analyzed first; results are used to informfollow-up qualitative data collection.

In what ways do the qualitative data help to explainquantitative findings?

Example: In what ways do the views of middle schoolboys with LD about their self-esteem explain whatthey reported about their self-esteem on surveys?

Exploratory Design: Qualitative data are collectedand analyzed first, results are used to informfollow-up quantitative data collection.

In what ways do the quantitative results generalizethe qualitative findings?

Example: Are the views of middle school boys withLD about their self-esteem generalizable to manymiddle school boys with LD?

Embedded Design: Qualitative and quantitative datacan be collected sequentially, concurrently, or both.One form of data is embedded within another.

How do the qualitative findings provide an enhancedunderstanding of the quantitative results?

Example: What are the characteristics of middleschool boys with LD who scored very high or verylow on self-esteem surveys?

Transformative Design: The researcher fi"ames thestudy within a transformative theoretical perspec-tive. Qualitative and quantitative data can be col-lected concurrently, sequentially, or both.

How do the qualitative findings provide an enhancedunderstanding of the quantitative results in orderto explore inequalities?

Example: How do the views of middle school boys withLD help researchers to develop a treatment program?

Multiphase Design: Both sequential and concurrentstrands are included in a study over a period of time(e.g., in a large-scale evaluation).

Include combinations of the previous questions atdifferent phases in the project so that an overallresearch goal is addressed.

Underlying Philosophical Concepts

In this section we describe different philosophical conceptsthat are important for understanding and framing mixed-methods research: paradigms, pragmatism, ontologies, andepistemologies. These concepts lay the groundwork forhow researchers view tmth, knowledge, and their work andexplain in part why different researchers value one approachto research over others. We include an illustration of howtensions between these concepts can play out while con-ducting special education research.

Paradigms. In his influential book The Structure of Sci-entific Revolutions (1962), Thomas Kuhn introduced theidea of "paradigm" as a specialized research term to meana worldview and the various philosophical assumptionsassociated with that point of view (Teddlie & Tashakkori,2009). Since then, a research paradigm has come to sig-nify a research culture that includes a set of beliefs, val-ues, and assumptions shared by a community of researchers(Johnson & Onwuegbuzie, 2004). These include ontologi-cal beliefs and epistemological beliefs, which we explainin a later section.

Researchers generally align with one of three researchparadigms (Teddlie & Tashakkori, 2009): (a) a positivist orpost-positivist paradigm (quantitative researchers), (b) a con-structivist paradigm (qualitative researchers), or (c) a prag-matist paradigm (mixed-methods researchers). Positivist andconstructivist paradigms predate the pragmatist paradigm.Creswell and Piano Clark (2011) noted that it was not untilthe 1980s that several researchers first described their use ofmixed methods. Thus, mixed-methods research is referredto as the third wave or third research movement (Johnson &Onwuegbuzie, 2004) and the "third research community"(Teddlie & Tashakkori, 2009, p. 4).

Pragmatism is a paradigm that "rejects the either/orchoices associated with the paradigm wars, advocates for theuse of mixed methods in research, and acknowledges thatthe values of the researcher play a large role in interpretationofresults" (Tashakkori & Teddlie, 2003, p. 713). Pragmatismfocuses on real life research problems and prioritizes thepurpose of the study rather than the use of particularresearch designs. Researchers mix methods in ways thatmake the most sense given their research questions.

KJingner and Boardtnan 211

integrating qualitative and quantitative approaches anddata analysis procedures and attempting "to open upinquiry to all possibilities while tying that search to practi-cal ends" (Maxcy, 2003, p. 86).

In describing the roots of pragmatism, Maxcy (2003)noted that it is a "distinctly American philosophy" (p. 56).He explained that pragmatism began as a philosophicalmovement during the late 19th century, initiated by CharlesSanders Peirce and developed fiarther by William James, JohnDewey, George Herbert Mead, and Arthur Bentley, amongothers. These early pragmatists challenged the perspectivethat social science research could access the real world solelythrough a single scientific method. They redirected "thecourse of philosophy away from continental idealism andNew World romantic transcendentalism and toward com-monsense practical thinking" (Maxcy, 2003, p. 55).

We are pragmatists. We identify with the early pragma-tists and their desire to address social problems and resolveconflicts. Maxcy (2003) explained that for them, "meaning-ful research began not with a single method or set of meth-ods but rather with ordinary experience and the desire for abetter world" (p. 53).

The parable of the elephant examined by several peoplein the dark (see Note 1 ) resonates with us and helps explainour preference for mixed-methods research. In this parable,each person feels a distinct, small part of the elephant, draw-ing a different conclusion about its nature. They then advo-cate for the veracity of their own conclusions, not realizingthat the reality is actually more expansive than they wereable to perceive individually. It was only possible to derivean accurate portrayal of the elephant by combining theirdiverse perspectives. As mixed-methods researchers, we areguided by constmctivism, sociocultural perspectives, andcognitive psychology, to varying degrees. We think in termsof continua rather than dichotomies. We conduct random-ized controlled trials, but believe that without the thoughtfulintegration of qualitative data, such quantitative studies cantell only part of the story (or help us feel part of the prover-bial elephant).

Ontologies and epistemologies. Ontology refers to fundamen-tal views about the nattire of reality. Quantitative researchersare often characterized as thinking of reality as objective andsingular. Qualitative researchers, on the other hand, think ofreality as subjective and multiple. Epistemology is closelyrelated to ontology and pertains to views of knowledge andwhere knowledge comes fi'om, how knowledge is acquired,whether something can be known beyond the shadow of adoubt, and what counts as knowledge (Creswell, 2008).What individuals think they know and the kinds of knowl-edge they value are based on epistemological assumptions.As Krauss (2005) explains, in the positivist paradigm, theobject of study is believed to be independent of researchers;researchers discover and verify knowledge through directobservations or measurements of phenomena, facts andestablish by taking apart a phenomenon to examine

its component parts. Altematively, the constructivist viewrepresents the perspective that knowledge is establishedthrough the meanings attached to the phenomena studied,researchers interact with the participants in a study to obtaindata, inquiry changes both researcher and participants, andknowledge is context and time dependent.

Researchers who adopt pragmatism are not necessarilyinterested in attempting to sort out epistemological/onto-logical issues. Rather, their interest lies in doing research thatyields useful results (or results that work). They embracevarious methodologies and perspectives. Greene (2005)focused on diversity and the melding of different episte-mologies. She described mixed methods as "an approach. . . that actively includes, even welcomes, multiple meth-odological traditions, multiple ways of knowing and multi-ple value stances" (p. 208). She added.

Ways of knowing are understood as also ways of valuing.A mixed method way of thinking thereby actively engageswith difference and diversity, again, of method and of val-ues, as well as difference and diversity as manifest in thecontexts in which we work. In short, a mixed method wayof thinking is itself anchored in values of toleration, accep-tance and respect—of multiplism and of difference. . . .Moreover, a mixed method way of thinking seeks better,more comprehensive understanding of educational phe-nomena, understanding that is woven from strands of par-ticularity and generality, contextual complexity andpattemed regularity, inside and outside perspectives, thewhole and its constituent parts, change and stability, equityand excellence and so forth. That is, a mixed method wayof thinking seeks not so much convergence as insight; thepoint is not a well-fitting model or curve but rather thegeneration of important understandings and discemmentsthrough the juxtaposition of different lenses, perspectives,and stances; in a good mixed methods study, difference isconstiUitive and fundamentally generative, (p. 208)

Spillane et al. (2010) cautioned mixed-methods researchersnot to lose touch with the particular ontological orepistemological fijndamentals of either qualitative or quan-fitative research. They also noted that although quantitativeand qualitative forms of research can be compatible, thechallenges of mixing these approaches should not be under-stated. Similarly, Johnson and Qnwuegbuzie (2004) wrotethat "philosophical debates will not end as a result of prag-mafism" (p. 17).

An example of converging epistemologies from specialeducation research: During a presentation at the 2010 PacificCoast Research Conference, Doug Fuchs summarized con-trasting ontologies and epistemologies for the audience. Heexplained that some scholars believe that the tmth about"what works" can be determined through hard work and rig-orous, carefully planned and executed research studies. Hethen noted that other scholars believe that tmth is relative


and varies depending on circumstances and on the perspec-tives of those involved. He was contrasting two prevalentresearch paradigms considered by many researchers to beopposing: a positivist (or post-positivist) paradigm and aconstmctivist paradigm. Positivists are aligned with the sci-entific method and believe that "there's a reality out there tobe discovered" (Krathwohl, 2009, p. 620). Constmctivistsbelieve that knowledge is subjective, shaped by one's expe-riences and background (Creswell, 2008).

Doug went on to tell about his Peer Assisted LeamingStrategies (PALS) Scaling Up project, explaining that theyhad three sites, one in Nashville, another in Minnesota, and athird in South Texas. The scale-up study began as a random-ized controlled trial with Grade 2-5 teachers assigned ran-domly to PALS and control groups. PALS teachers leamedPALS and were provided with support to help them imple-ment the program correctly. Findings across sites were mixed.Also, it appeared that adding "helpers" to boost implementa-tion fidelity did not lead to greater outcomes (McMasteret al., 2010). McMaster et al. wrote, "The impact of an'evidence-based' intervention may vary with changes in edu-cational contexts" (p. 2).

Doug explained that they changed their design in the nextyear of the study to provide teachers with the option of makingadjustments to PALS (but keeping certain critical componentsintact). Controls continued as controls; however, PALS teach-ers were able to choose whether they wanted to continue as"top-down" teachers conducting PALS "by the book" or makechanges and be "boftom-up" teachers. Thus, in Year 2, thestudy became quasi-experimental because teachers were notassigned randomly to the two PALS conditions. McMaster etal. (2010) compared the achievement of students taught by top-down PALS teachers with the achievement of students taughtby boftom-up PALS teachers and found that students who usedmodified versions of PALS outperformed their peers.McMaster et al. concluded, "It seems that teachers' customiza-tions increased PALS 'fit' with their specific classroom needs."It appeared that context mattered. When we wrote to Dougabout this, he explained that they looked hard for altemativeways of explaining their findings, such as possible differencesbetween top-dovra and boftom-up teachers and their students,but could not find any. He cautioned, "But that doesn't meanthey don't exist... So, 'context mafters' is a logical conclusionto draw from our findings, but it is not the only possible con-clusion" (D. Fuchs, personal communication, October 29,2010). We also asked Kristen McMaster about this research(K. McMaster, personal communication, October 29, 2010),and she shared the following:

Evidence suggests that a balance of fidelity to core compo-nents of PALS and flexibility to customize PALS to fit par-ticular classroom and student needs may enhance studentachievement outcomes. In future research, it will be impor-tant to carefully consider what types of data and methodolo-gies will shed light on relations among fidelity, flexibility,and student outcomes. Additional data could include class-

room observations and teacher and student interviews toelucidate the extent to which teacher buy-in, student moti-vation and engagement, and how other contextual vari-ables influence teachers' implementation of evidence-basedpractices and student gains that result.

After the PCRC presentation, we asked Doug whichapproach he now believes is best, and he responded thathe thinks the answer is "somewhere in the middle." Webelieve that "somewhere in the middle" is a good way todescribe mixed-methods research. In addition, we think thatthis illustration shows why it is so valuable to include quali-tative data when conducting intervention studies. Given thecultural and linguistic diversity of their sample, it also dem-onstrates the need for more research that focuses on lan-guage and culture.

Culture and Mixed Methodsin Special Education Research

To make progress toward closing the research gap wedescribed at the beginning of this article, we believe thatspecial education researchers must broaden their perspec-tives by first understanding research as situated culturalpractice and then conducting more mixed-methods researchthat explicitly addresses the needs of culturally and linguis-tically diverse students (Arzubiaga, Artiles, King, & Harris-Mun-i, 2008).

Research on culturally and linguistically diverse populations.Researchers have not done enough to focus on culturally andlinguistically diverse students. The assumption by many seemsto be that "one size fits all," or, in other words, that the lessonsleamed from research with mostly White, middle class stu-dents can be applied to other students. For example, theNational Reading Panel report "did not address issues relevantto second language leaming" (National Institute of ChildHealth and Human Development, 2000, p. 3). Yet the recom-mendations from the report were widely touted as applying toall students (i.e., through Reading First). On the other hand, theNational Early Literacy Panel (2008) did look for studies con-ducted with English language leamers, as well as other cultur-ally diverse children, but noted that there were "not yet studiesfocusing on these specific subpopulations or that allow exami-nation of these subpopulations" (p. 120). Despite a lack ofresearch, the authors recommended that it would be "pmdent"to make a practice found to be effective with monolingualEnglish-speaking children, code-focused instmction, "avail-able to all populations of young children, at least until researchmore directly addresses this question" (p. 120). However, theextent to which code-based instmction helps young Englishlanguage leamers is an empirical question and one that we sug-gest would best be answered through mixed-methods research.Other research suggests that a more pmdent approach withEnglish language leamers would actually be to focus on orallanguage skills, vocabulary, and background knowledge


(Dickinson, Hirsh-Pasek, Neuman, Btirchinal, & Golinkofï,2009; Gutierrez, Zepeda, & Castro, 2010).

Prichard and Klingner (2010) examined reading interven-tion studies published between 2001 and 2009 and found thatin several research studies, the authors did not give adequateconsideration to student variation (e.g., regarding ethnicity andlanguage proficiency) and overgeneralized their findings asapplying to students not part of their research populations.Similarly, Artiles, Trent, and Kuan (1997) conducted a reviewof 22 years of empirical research published in special educa-tion journals and found that diversity was rarely addressed.They noted that the number of articles published with a focuson diversity was "alarmingly low" and concluded that manyeducational researchers form generalizations about culttirallyand linguistically diverse students without attending to popula-tion validity and ecological validity or designing research thatis sensitive to cultural differences. Lindo (2006) analyzedinterventions for AfHcan American student populations andfound few studies that specifically examined and reporteddemographic results. In analyzing research conducted withEnglish language learners. Bos and Fletcher (1997) and Artileset al. (2005) found a scarcity of research on within-groupdiversity among English language leamers. Researchers rarelydescribed in sufficient detail the demographics, language pro-ficiency, and other characteristics of English language leamers.Without this information it is difficult to determine the validityof an intervention for different subpopulations of students. Werecommend much greater attention to understanding the char-acteristics of participants in research studies.

A related problem is that English language leamers areoften omitted from research samples precisely because oftheir less-than-fluent language. Yet language dominance andproficiency are important research variables and can and doaffect treatment outcomes. Leaving students out of studiesñirther limits the extemgj validity and applicability of suchstudies, especially for teachers who have English languageleamers in their classes.

Research as situated cultural practice. One reason that moreresearchers do not focus on cultural and linguistic diversitymay be that they operate within a post-positivist paradigmand a belief that research and practice can be acultural. Bydisregarding the infiuence of culture, they view research,as well as teaching and leaming, as technical enterprises(Arzubiaga et al., 2008). They do not understand the culturalnature of research or adequately account for culture whenconceptualizing, designing, and carrying out their investiga-tions. In emphasizing the importance and centrality of cul-ture when conducting research in the social sciences,Demerath (2006), citing Steams (2003), noted that "cultureinfluences, indeed, powerñilly shape the human condition."More systemic inquiry on cultural processes "could ele-vate the place of cultural findings in an intellectual com-munity that, particularly in the United States, has becomeexcessively devoted to a scientism that tends to ignoreculture" (p. 101).

Thus, we argue for a paradigm shift so that research willbe seen as situated cultural practice. Arzubiaga et al. (2008)challenged special education researchers to view themselvesas cultural beings and members of a scientific field, and torecognize the cultural presuppositions in that field's habitualpractices. What problems to study, which theoretical per-spectives to adopt, where to conduct research and with whom,which measures to use, how to analyze data, which findingsto highlight, and how to interpret results are all informed bycultural presuppositions. Acknowledging this is perhaps thefirst step toward becoming more culturally aware. AsArzubiaga et al. noted, "Special education and psychologyresearch must be based on a view of 'human nature thatplaces culture at the center of its concerns' (Cole, 1998,p. 291)" (p. 312).

The second step is to move beyond a view of culture as astatic, unitary constmct that is "neatly bound across groups"(Arzubiaga et al., 2008, p. 312) or that stereotypes individu-als and boxes them into categorical identities (e.g., poor.Latino, low-achieving student). Rather, culture is complexand dynamic (Gutierrez & Rogoff, 2003; Rogoff, 2003).Culture reflects one's beliefs, how one learns, what onevalues, and how one interacts with others (Klingner &Soltero-González, 2009). Within any cultural group is muchvariability. Individuals belong to multiple cultures into andout of which they move fluidly. Cole and Engeström (1993)explained that although cultures demonstrate pattems, "thereis also no doubt that [they are] far from uniform, because[they are] experienced in local, face-to-face interactions thatare locally constrained and heterogeneous with respect toboth 'culture as a whole' and the parts of the entire toolkitexperienced by a given individual" (p. 15). It is important toconsider the cultural practices an individual brings to a situ-ation, as well as the institutional cultures that provide thecontext within which interactions take place (McDermott,Goldman, & Varenne, 2006). Arzubiaga et al. (2008) notedthat adopting this view of culture as cultural practices allowsresearchers to pose more complex questions about leamingand instmction than otherwise would be possible. It is throughthese complex questions, addressed with mixed methods, thatresearchers can begin to close the research gap.

An example of the power of qualitative research to explainquantitative findings: Success for All (SFA) has been a highlytouted instmctional program. For example, it was one ofthree school reform programs found to be strongly associatedwith improvement in student achievement by the AmericanInstitutes for Research (Herman et al., 2000). Yet, studies ofthe effects of a 3-year implementation of SFA in Miami-Dade County, Florida, showed that achievement did notimprove in the highest-need schools (Levitt, 2000) and thatEnglish language leamers in SFA schools made smaller gainsin English proficiency than students in comparison schools(Urdegar, 2000). The creators of SFA wamed that their pro-gram is effective only when "ftjlly implemented" (Slavin &Madden, 2001, p. 34), and observers commented that SFA


must not have worked in Miami-Dade County because itwas not implemented with fidelity.

As part of a larger study in South Florida, schools focusedon understanding the reasons for the disproportionate repre-sentation of culturally and linguistically diverse students inspecial education (Harry & Klingner, 2006). Klingner,Cramer, and Harry (2006) examined the challenges faced byfour high-need urban schools when trying to implement SFA.The students in these schools were predominantly AfricanAmerican or Caribbean Black (mostly from Haiti), withalmost 100% of the students receiving free or reduced-pricelunches. Klingner and colleagues wanted to understand thereading instmction students were receiving as well as howSFA fit into the larger school context. Over a span of almost2 years, the researchers observed 45 SFA lessons (21 com-plete and 24 partial) across the four schools, implemented by30 different teachers. Klingner et al. found that several issuesaffected the fidelity with which teachers implemented SFA.One of the greatest challenges was that students were not ableto pass the tests that would allow them to move to higherlevels of the program. Thus, they recycled through materialthey had already covered, sometimes many times. This addedto students' boredom and fmsfration and created behavioralchallenges. Teachers faced numerous grouping and schedul-ing difficulties, resulting in students of different reading lev-els and widely varying grade levels being placed in the sameclass. It appeared that teachers tried in good faith to imple-ment SFA as it was designed but faced very real challengesthat made it difficult to do so. Perhaps most importantly, itwas the teachers who made adaptations so that SFA couldwork in their classrooms, with their students, who seemed tobe most effective. By observing in classrooms and interview-ing teachers, Klingner et al. leamed a great deal that couldinform future efforts to implement SFA in high-need urbanschools. These rich qualitative findings are important whenviewed in light of the quantitative outcomes, namely, lowfidelity and low student scores. It is qualitative methods thatare most responsive to the local context. This example showswhy a combination of qualitative and quantitative methods isbest able to answer questions about complex phenomena. Italso illustrates the importance of understanding the contextwithin which interventions are carried out, as well as culturalpractices at play. Additionally, it demonsfrates what Ball andForzani (2007) described when they noted, "Ofren lacking isresearch that explains causes or examines the interplay at theheart of educational practice and policy." As do we, theyargued for research that focuses on the "instmctionaldynamic" (p. 529).

Debunking Medical Scienceas the Gold Standard in Research

We contend that the research gap in special education is duein part to a narrow view of what counts as research. Several

years ago, the first author sat in the audience at an Office ofSpecial Education Programs Project Directors conference asRuss Whitehurst (then the Director) explained that the fieldof education has much to leam from the field of medicine.He held up randomized controlled trials as the gold standardfor research. However, medicine is not the exact science it issometimes thought to be. When the first author was freatedfor a relatively rare form of cancer a few years ago and readevery article she could find on her condition in medicaljoumals and talked with doctors about a treatment plan, itbecame very clear to her that randomized controlled trialscould provide little guidance. Not enough research hadbeen conducted on her disease because there were too fewpatients with her particular form of cancer. No treatmenthad been found to have a statistically significant positiveeffect, in part because sample sizes were too small. Whenshe asked about effect sizes, her doctors had no idea whatshe meant. Trying to follow recommendations from researchconducted with patients with related conditions seemed oflittle help because each doctor with whom she spoke had adifferent opinion of what the best course of action might be.Ultimately, the treatment plan they came up with seemed tohave as much to do with what they determined about her,her life style, and unique aspects of her condition as it didwith guidance from the results of clinical trials. In otherwords, her doctors used mixed-methods research to formu-late a plan, and her role was an active rather than passiveparticipant.

Mixed methods are more prevalent in medicine thanone might realize. Patients respond differently to the sametreatment for many reasons, some of which medical expertsindicate seem to be related to race (Bloche, 2004;Committee on Pharmacokinetics and Dmg Interaction inthe Elderly, Institute of Medicine, 1997; Harder, 2005).Yet, these differences very well could be due to social ineq-uities and disparities in the health care provided across eth-nic groups (Smedley, Stith, & Nelson, 2003).

Regardless of race, the effectiveness of medical freat-ments varies depending on multiple factors. Just as all stu-dents do not leam in the same way, not all patients respond inthe same way to medical treatments. Consider medicationsfor depression (Klingner, Sorrells, & Barrera, 2007). Forexample, although Paxil has been found to be effective forreducing depression through clinical friáis with adults, amongadolescents Paxil actually increases incidences of suicide andwas banned in Great Britain (Alliance for Human ResearchProtection, 2003). Even with adults, Paxil, like other medi-cations for depression, seems to work well for some patientsbut not others. The process of identifying the most effectivedmg for a given individual seems to be one of frial and errorand collecting multiple sources of data, including interviewswith the patient. Decisions about the most effective treat-ment are based on a variety of influences, forces, anddecisions. Even in the field of medicine, it is not enough


to ask "What works?" Rather, physicians must ask, "Whatworks with whom, under what circumstances, and withwhat effects?"

Erickson (2005) also challenged the idea of randomizedfield trials as the gold standard for educational research bydrawing from the field of medicine. He relayed how a phy-sician, upon hearing this idea during a discussion about theNafional Research Council's (Shavelson & Towne, 2002)report on educational research, commented that "if knowl-edge development in polio research had had to depend onlyon conclusive findings from experiments, research on poliowould today consist mainly of studies of the treatment effectsof the iron lung" (p. 9). Just as in medicine, in education thereis a place for large-scale clinical trials. But, as Erickson,noted, "the questions such approaches know how to answerare not the only ones worth asking" (p. 9). The potentialpower of qualitative data to enhance and explain quantitativefindings should not be undervalued. We argue that the newgold standard in special education should be mixed-methodsresearch that combines the best of multiple approaches inorder to answer the "questions worth asking."

A Mixed Methods ExampleFronn Our Research:Collaborative Strategic Reading

In this section, we highlight an example of mixed methodsfrom our current research on Collaborative Strategic Reading(CSR) with stmggling readers. The quantitative findingsfrom this randomized control trial research are reported inVaughn et al. (in press; see Note 2). CSR is an instmcfionalapproach designed to enhance the reading comprehensionand content leaming of students in diverse, heterogeneousclassrooms. Students leam reading comprehension strate-gies and apply them while working with their peers in small,student-led collaborative groups (Klingner, Vaughn, Dimino,Schumm, & Bryant, 2001). To investigate the complexrelationships between teaching reading strategies and studentleaming outcomes, we chose a sequential mixed-methodsexplanatory design (Creswell & Piano Clark, 2011;Onwuegbuzie & Teddlie, 2003; Tashakkori & Teddlie,1998) to formulate our research questions and to guide ourdata collection and analysis. We focus here on the processof creating research questions and the progression of datacollection related to student outcomes.

In our sequential mixed-methods research, we used researchquestions as a guide in an iterative process in which we ana-lyzed data while we confinued to collect addifional data.Initial findings informed subsequent quesfioning, data col-lection, and analysis. For example, in our first overarchingquestion, we asked: "Does Collaborative Strategic Readingimprove reading outcomes for stmggling readers in generaleducation language arts and reading classrooms?"

We chose inifial data-collecfion procedures by establish-ing the kinds of data needed to answer this research question.This included providing pre- and postassessments to all stu-dents in intervention and comparison classes. With this ini-tial quantitative data analysis we could determine if CSRwas working and for whom. We assessed the significance ofthe effect (p value) and the magnitude (effect size) and cor-related these results with student demographic information.With results from these quanfitative analyses, we were ableto ask a second research question: "What components ofCSR contribute to student leaming?"

To answer this research question, we began with our quan-titative results and then determined how we could probemore deeply into students' leaming. We colleeted addi-tional data for analysis. We selected a representative sam-ple of students for each teacher in the study (i.e., one classof students for each teacher). These students were providedwith qualitative measures that included the analysis of stu-dent leaming logs (student work products) and a discourseanalysis of audio-reeorded group discussions during classto detennine how the strategies were enacted in the class-rooms. This pronged approach provided a rich descriptionof how the strategies were used by students and allowed usto correlate and triangulate findings across multiple datasources. In some cases, such as with student work products,qualitative data were "quantitized" (converted into quanti-tative representations; Tashakkori & Teddlie, 1998) in orderto interpret results. The quantitative and qualitative datarevealed themes about how strategies were used as well asthe variability in the ways that students seemed to take upthe use of new strategies. From these combined results, wesee that additional data are needed to probe the use of read-ing strategies with stmggling readers. Thus, we are now ask-ing a third question: "How do stmggling readers who madereading gains differ from stmggling readers who did notleam reading strategies?"

To answer this question, we plan to formulate groups ofstudents, on the basis ofresults from quantitative and quali-tative findings, who either improve in strategy use or do notimprove in strategy use during the study. A sample of thesestudents will participate in an additional layer of data col-lection. These students will be given a think-aloud task toassess strategy use and will also be interviewed to deter-mine their perceptions of the new reading strategies.Additional students will be interviewed and provided thethink-aloud task until saturation is achieved (i.e., no newinformafion emerges during coding; Strauss & Corbin,1998). The integration of quanfitative data from assessmentscores with qualitative data from interviews, think-alouds,and leaming logs should allow for an in-depth understand-ing of how struggling readers use reading strategies andwhy some stmggling readers improve more than others.

Answering the question "What works?" is only one com-ponent of our CSR research. Through mixed methods, we


are able to probe more deeply into the complexities of howand why CSR might be associated with improved leamingoutcomes for stmggling readers. We can better understandthe circumstances under which CSR seems to be most effec-tive, and with whom.

Conclusion

We have argued that there is a research gap in special edu-cation at least in part because the field has not embracedmultiple research methodologies better suited to answeringa range of important questions than quantitative-onlydesigns. While experimental research provides a powerfuland rigorous tool for answering questions about the effi-cacy of instmctional methods, it does not readily addressquestions related to how and why these methods work orunder what circumstances (Collins, Onwuegbuzie, &Sutton, 2006). In particular, it is not well-suited for address-ing the complex issues found in today's culturally, linguisti-cally, and socio-economically diverse classrooms.

Mixed-methods research has the potential to addressreal world issues and to provide the "optimal combinationrequired for the powerful development of evidence and anexplanation that will gain a consensus around the interpreta-tion of the data" (Krathwohl, 2009, p. 620). Mixed-methodsresearch has evolved to the point that it is a separate meth-odological orientation with its own worldview, vocabulary,and techniques (Tashakorri & Teddlie, 2003). It draws fromquantitative and qualitative research traditions and com-bines them in unique ways to answer questions that cannotbe answered through other means.

Yet mixed-methods research alone cannot solve everyproblem in education. Different approaches are needed. Allresearch methods have strengths and weaknesses, and it isimportant to understand what each research methodologycan and cannot do. Johnson and Onwuegbuzie's (2004) posi-tion reflects our thinking:

We do not believe that mixed-methods research is currentlyin a position to provide perfect solutions. Mixed-methodsresearch should, instead (at this time), use a method andphilosophy that attempt to fit together the insights providedby qualitative and quantitative research into a workablesolution, (p. 16)

A possible weakness with mixed-methods research is that itdoes not automatically position research as situated culturalpractice or account for the cultural practices inherent in school-ing (Arzubiaga et al., 2008). Thus, we believe that researchersmust go beyond mixed methods and a pragmatist stance toembrace a socioculttiral perspective. A mixed-methods para-digm allows the space and flexibility to do this.

One crificism of some mixed-methods studies is that theresearchers do not sufficiently integrate their qualitativeand quantitative data. Spillane et al. (2009) lamented that

mixed-methods researchers "often use the two approaches inparallel, rather than in tandem. As a result, the potential ofmixing methods is not maximized" (p. 7). Onwuegbuzie,Slate, Leech, and Collins (2009) asserted that the data analy-sis step can make or break a mixed-research study. Theyclaimed that the more interactive and embedded the quantita-tive and qualitative analyses in a mixed methods study are,the more integrated and coherent the inferences form thestudy will be.

Mixed-methods research is a relatively new paradigm,with untapped potential. We hope that our colleagues in spe-cial education will embrace it as a way to make progresstoward solving some of the field's most intractable problems.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respectto the research, authorship, and/or publication of this article.

Funding

The authors received the following financial support for the research,authorship, and/or publication of this article: The CollaborativeStrategic Reading research described in this article is supported byGrant R305A080608 from the Institute of Education Sciences, U.S.Department of Education. The content is solely the responsibility ofthe authors and does not necessarily represent the official views of theInstitute of Education Sciences or the U.S. Department of Education.

N o t e s

1. There are various versions of this parable. In some versions, theindividuals are blind rather than in the dark. For one source forthis parable, see Saxe, J. G. (1963). The blind men and the ele-phant: John Godfrey Saxe's version of the famous Indian legend.Pictures by Paul Galdone. New York: Whittlesey House.

2. We included findings from qualitative data in an earlier draftof this manuscript, but a reviewer and the joumal editor askedus to delete them.

References

Alliance for Human Research Protection. (2003, June 10). UKhealth department/British prime minister issue statementwarning no Paxil for children! Retrieved from http://www.ahrp.org/infomail/0603/10.php.

Artiles, A. J., Rueda, R., Salazar, J. J., & Higareda, I. (2005).Within-group diversity in minority disproportionate repre-sentation: English language leamers in urban school districts.Exceptional Children, 71, 283-300.

Artiles, A. J., Trent, S. C, & Kuan, L. A. (1997). Leaming disabilitiesresearch on ethnic minority students: An analysis of 22 years ofstudies published in selected refereed joumals. Leaming Dis-abilities Research & Practice, 12, 82-91.

Arzubiaga, A. E., Artiles, A. J., King, K. A., & Harris-Murri, N.(2008). Beyond research on cultural minorities: Challenges andimplications of research as situated cultural practice. Excep-tional Children, 74, 309-327.


Ball, D. L., & Forzani, F. M. (2007). 2007 Wallace Foundationdistinguished lecture: What makes education research "educa-tional"? Educational Researcher, 36, 529-540.

Berliner, D. C. (2002). Educational research: The hardest science ofall. Educational Researcher, i7(8), 18-20.

Bloche, M. G. (2004). Race-based therapeutics. New EnglandJournal of Medicine, 351, 2035-2037. Extract available athttp://content.nejm.org/cgi/content/extract/351/20/2035.

Bos, C. S., & Fletcher, T. V. (1997). Sociocultural considerationsin leaming disabilities inclusion research: Knowledge gaps andfuture directions. Learning Disabilities Research & Practice,12, 92-99.

Chatterji, M. (2005). Evidence on "what works": An argumentfor extended-term mixed-method (ETMM) evaluation design.Educational Researcher, 34(5), 14-24.

Cole, M. (1998). Can cultural psychology help us think about diver-sity? Mind, Culture, and Activity, 5,291-304.

Cole, M., & Engeström, Y. (1993). A cultural-historical approachto distributed cognition. In G. Salomon (Ed.), Distributedcognitions: Psychological and educational considerations(pp. 88-110). New York, NY: Cambridge University Press.

Collins, K. M. T, Onwuegbuzie, A. J., & Sutton, I. L. (2006).A model incorporating the rationale and purpose for conduct-ing mixed-methods research in special education and beyond.Learning Disabilities: A Contemporary Journal, 41, 67-100.

Committee on Pharmacokinetics and Drug Interaction in theElderly, Institute of Medicine (1997). Pharmacokinetics anddrug interactions in the elderly and special issues in elderlyAfrican-American populations: Workshop summary. Washing-ton, DC: National Academies Press.

Creswell, J. W. (2008). Research design: Qualitative, quantitative,and mixed methods approaches (3rd ed.). Thousand Oaks,CA: Sage.

Creswell, J. W., & Piano Clark, V. L. (2011). Designing and con-ducting mixed-methods research (2nd ed.) Thousand Oaks,CA: Sage.

da Silva, C. D., Huguley, J. P., Kakli, Z., & Rao, R. (2007). Theopportunity gap: Achievement and inequality in education.Cambridge, MA: Harvard Educational Review.

Demerath, P. (2006). The science of context: Modes of responsefor qualitative researchers in education. International Journalof Qualitative Studies in Education, 19, 97-113.

Dickinson, D., Hirsh-Pasek, K., Neuman, S., Burchinal, M., &Golinkoff, R. (2009). The language of emergent literacy: Aresponse to the National Institute for Literacy report on earlyliteracy. New Brunswick, NJ: National Institute for EarlyEducation Research.

Donovan, S., & Cross, C. (2002). Minority students in specialand gifted education. Washington, DC: National AcademyPress.

Eisenhart, M. (2006). Qualitative science in experimental time.International Joumal of Qualitative Studies in Education,19(6), 697-707.

Elmore, R. (1996). Getting to scale with good educational practice.Harvard Educational Review, 66(\), 1-26.

Erickson, F. (2005). Arts, humanities, and sciences in educationalresearch and social engineering in federal education policy.Teachers College Record, 107(\), 4-9.

Feuer, M. J., Towne, L., & Shavelson, R. J. (2002). Scientific cultureand educational research. Educational Researcher, J/, 4-14.

Gee, J. P. (2001). A sociocultural perspective on early literacydevelopment. In S. B. Neuman & D. K. Dickinson (Eds.),Handbook of early literacy research (pp. 30-42). New York:Guilford Press.

Greene, J. (2005). The generative potential of mixed methodsinquiry. International Journal of Research á Method in Edu-cation, 2 8,201-2\\.

Gutierrez, K. D., & Rogoff, B. (2003). Cultural ways of leam-ing: Individual traits of repertoires of practice. EducationalResearcber, 32(5), 19-25.

Gutierrez, K. D., Zepeda, M., & Castro, D. C. (2010). Advancingearly literacy leaming for all children: Implications of the NELPreport for dual language leamers. Educational Researcher, 39,334-339.

Harder, B. (2005). The race to prescribe: Drug for AfricanAmericans may debut among debate. Science News, 167,247-248.

Harry, B., & Klingner, J. K. (2006). Why are so many minority stu-dents in special education? understanding race and disabilityin schools. New York, NY: Teachers College Press.

Herman, R., Carl, B., Lampron, S., Sussman, A., Berger, A., &Innes, F. (2000). What we know about comprehensive schoolreform models. Washington, DC: American Institutes forResearch.

Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed-methodsresearch: A research paradigm whose time has come. Educa-tional Researcher, 33(1), 14-26.

Johnson, R. B., Onwuegbuzie, A. J., & Tumer, L. A. (2007).Toward a definition of mixed-methods research. Journal ofMixed-methods Research, 1(1), 112-133.

Klingner, J. K., Cramer, E., & Harry, B. (2006). Challenges inthe implementation of Success for All by four urban schools.Elementary School Journal, 106, 333-349.

Klingner, J. K., & Edwards, P. (2006). Cultural considerations withresponse to intervention models. Reading Research Quarterly,41, 108-117.

Klingner, J., & Soltero-González, L. (2009). Culturally and lin-guistically responsive literacy instruction for English languageleamers with leaming disabilities. Multiple Voices, 12(\), 1-17.

Klingner, J. K., Sorrells, A. M., & Barrera, M. (2007). Three-tieredmodels with culturally and linguistically diverse students. InD. Haager, J. Klingner, & S. Vaughn (Eds.), Evidence-basedpractices for response to intervention (pp. 223-244). Baltimore,MD: Brookes.

Klingner, J. K., Vaughn, S., Dimino, J., Schumm, J. S., &Bryant, D. (2001). Collaborative Strategic Reading:


Strategies for improving comprehension. Longmont, CO:Sopris West.

Krathwohl, D. R. (2009). Methods of educational and social sci-ence research. Long Grove, IL:'Waveland Press.

Krauss, S. E. (2005). Research paradigms and meaning making: Aprimer. The Qualitative Report, ¡0(4), 758-770.

Lee, J. (2002). Racial and ethnic achievement gap trends: Revers-ing the progress toward equity. Educational Researcher, 31,3-12.

Levitt, J, (2000), An interim evaluation of operation safety net:Three-year report. Miami, FL: Miami-Dade County PublicSchools, Otïice of Evaluation and Research,

Lindo, E, J, (2006), The African American presence in readingintervention research. Remedial and Special Education, 27,148-153.

Maxcy, S. J. (2003). Pragmatic threads in mixed-methodsresearch in the social sciences: The search of multiple modesof inquiry and the end of the philosophy of formalism. InA. Tashakkori & C. Teddlie (Eds.), Handbook of mixedmethods in social and behavioral research (pp. 51-89).Thousand Oaks, CA: Sage.

McDermott, R., Goldman, S., & Varenne, H. (2006). The culturalwork of leaming disabilities. Educational Researcher, 35(6),

. 12-17.McMaster, K. L,, Fuchs, D., Saenz, L., Lemons, C , Keams, D., &

Yen, L,, et al, (2010). Scaling up PALS: Importance of imple-menting evidence-based practice with fidelity and flexibil-ity. New Times for DLD, 28(\), 1-2, 4.

National Early Literacy Panel. (2008). Developing early literacy:Report of the National Early Literacy Panel. Washington, DC:National Institute for Literacy. Available at http://www.nifl,gov/earlychildhood/NELP/NELPreport,html

National Institute of Child Health and Human Development,(2000), Report of the National Reading Panel: Teachingchildren to read: An evidence-based assessment of the scien-tific research literature on reading and its implications forreading instmction: Reports of the sub-groups. Washington,DC: U.S. Department of Health and Human Services, NationalInstitute on Health.

Onwuegbuzie, A. J., Slate, J. R., Leech, N. L., & Collins, K. M. T.(2009). Mixed data analysis: Advanced integration techniques.International Journal of Multiple Research Approaches, 3,13-3.

Onwuegbuzie, A. J., & Teddlie, C. (2003). A framework for ana-lyzing data in mixed-methods research. In A. Tashakkori &C. Teddlie (Eds.), Handbook of mixed methods in social andbehavioral research (pp. 351-383), Thousand Oaks, CA: Sage,

Prichard, B, A,, & Klingner, J, K, (2010), Addressing the needs ofEnglish language learner populations: A synthesis of the pop-ulation validity of reading intervention research. Manuscriptsubmitted for publication,

Raudenbush, S. W. (2005). Leaming from attempts to improveschooling: The contribution of methodological diversity. Edu-cational Researcher, 34(5), 25-31.

Rogoff, B. (2003), The cultural nature of human development.New York: Oxford University Press,

Schneider, B,, & McDonald, S, K, (2007). Scale-up in educa-tion: Issues in practice (Vol. 2). New York, NY: Rowman andLittlefield,

Shavelson, R. J., & Towne, L. (Eds.), (2002). Scientific researchin education. Washington, DC: National Research Council,National Academy Press,

Slavin, R, E,, & Madden, N,A, (2001), Success for All and Com-prehensive School Reform: Evidence-based policies for urbaneducation. Washington, DC: Office of Educational Researchand Improvement,

Smedley, B. D,, Stith, A, Y, & Nelson (Eds,), (2003). Unequaltreatment: Confronting racial and ethnic disparities in healthcare. Washington, DC: National Academies Press.

Spillane, J.P, Pareja, A, S,, Domer, L,, Barnes, C , May, H., Huff, J.,& Cambum, E. (2010). Mixing methods in randomized con-trolled trials (RCTs): Validation, contextualization, triangula-tion, and control. Educational Assessment, Evaluation andAccountability 22, 5-28.

Steams, P. (2003, May) Expanding the agenda of cultural research.Chronicle of Higher Education, 2, p. 7.

Strauss, A., & Corbin, J. (1998). Basics of qualitative research(2nd edition). Thousand Oaks, CA: Sage.

Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Com-bining qualitative and quantitative approaches. Applied SocialResearch Methods Series (vol. 46). Thousand Oaks, CA: Sage.

Tashakkori, A., & Teddlie, C. B. (2003). Handbook of mixedmethods in social and behavioral research. Thousand Oaks,CA: Sage.

Teddlie, C , & Tashakkori, A. (2009). Foundations of mixedmethods research: Integrating quantitative and qualitativeapproaches in the social and behavioral sciences. ThousandOaks, CA: Sage.

Urdegar, S. (2000). Evaluation of the Success for All program:1998-1999. Miami, FL: Miami-Dade County Public Schools,Office of Evaluation and Research.

Vaughn, S., Klingner, J., & Hughes, M. (2000). Sustainabil-ity of research-based practices. Exceptional Children, 66,163-171.

Vaughn, S., Klingner, J. K., Swanson, E. A., Boardman, A. G.,Roberts, G., Mohammed, S. S., & Stillman-Spisak, S. J. (inpress). Efficacy of Collaborative Strategic Reading withmiddle school students, American Educational ResearchJournal.

About the Authors

Janette K. Klingner is a Professor of Bilingual MulticulturalSpecial Education at the University of Colorado at Boulder.Currently, her principal area of research focuses on reading com-prehension strategy instruction for culturally and linguisticallydiverse students.

Alison G. Boardman is an Assistant Research Professor in theSchool of Education at the University of Colorado at Boulder. Herresearch interests include interventions for struggling readers andproviding effective professional development and coaching toteachers.

COPYRIGHT INFORMATION

Author:

Title:

Source:

ISSN: DOI:

Publisher:

Klingner, Janette K.; Boardman, Alison G.

Addressing the "Research Gap" in Special Education Through Mixed

Learn Disabil Q 34 no3 Ag 2011 p. 208-218

0731-9487

10.1177/0731948711417559

Council for Learning Disabilities

11184 Antioch, Suite 405, Overland Park, KS 66210

The magazine publisher is the copyright holder of this article and it is reproducedwith permission. Further reproduction of this article in violation of the copyright isprohibited.

This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub- licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

addressing the ''research gap' in special education...

Documents